Search

With the continuous improvement of shale oil and gas recovery technologies and achievements, a large amount of geological information and data have been accumulated for the description of shale reservoirs, and it has become possible to use machine learning methods for "sweet spots" prediction in shale oil and gas areas. Taking the Duvernay shale oil and gas field in Canada as an example, this paper attempts to build recoverable shale oil and gas reserve prediction models using machine learning methods and geological and development big data, to predict the distribution of recoverable shale oil and gas reserves and provide a basis for well location deployment and engineering modifications. The research results of the machine learning model in this study are as follows: ① Three machine learning methods were applied to build a prediction model and random forest showed the best performance. The R2 values of the built recoverable shale oil and gas reserves prediction models are 0.7894 and 0.8210, respectively, with an accuracy that meets the requirements of production applications; ② The geological main controlling factors for recoverable shale oil and gas reserves in this area are organic matter maturity and total organic carbon (TOC), followed by porosity and effective thickness; the main controlling factor for engineering modifications is the total proppant volume, followed by total stages and horizontal lateral length; ③ The abundance of recoverable shale oil and gas reserves in the central part of the study area is predicted to be relatively high, which makes it a favorable area for future well location deployment. [ABSTRACT FROM AUTHOR]

The Chang 8 reservoir of the Huanjiang Oilfield in the Ordos Basin is a tight sandstone reservoir with poor reservoir physical properties and uneven oil distribution. In this study, the effective reservoirs developed on a large scale under the condition of horizontal well volume fracturing technology of the Chang 8 Member was identified based on the data of core observation, experimental analysis, logging, oil test, and production dynamics, and the identification standard of effective reservoirs in terms of reservoir physical properties, oil content, and comprehensive logging characteristics of gas logging was established. This scheme allows for a thorough identification of the effective reservoirs for horizontal well development in tight sandstone reservoirs. The findings indicate that the study area’s tight oil resources can be successfully produced. The well logging results show that the oil-bearing property of sand bodies with oil stains is better than that of oil spots. The bottom limit of the acoustic wave time difference is 210 μs/m, the permeability is 0.03 mD, the porosity is 6.0%, the rock resistivity is 30 Ω m, and the total hydrocarbon gas measurement value exceeds five times the baseline. At the same time, the total oil thickness is greater than 6 m, and the thickness of a single sand body is above 4 m. We have defined the lower limit standard of high-efficiency reservoir of Chang 8 tight oil in L289 block of Huanjiang oilfield. According to the analysis of oil-bearing property, comprehensive logging display, and reservoir thickness, the geological “sweet spots” is optimized to provide reference for subsequent mining. Through comparative analysis, the rules and trends are found to provide a basis for selecting mining strategies. With the help of technical means such as numerical simulation and geological modeling, the prediction accuracy and decision-making effect are improved. By clarifying the lower limit standard of reservoir, optimizing geological “sweet spots” and avoiding risk areas, the mining efficiency is improved and the cost is reduced, which provides reference for similar oilfield development.

With the continuous improvement of shale oil and gas recovery technologies and achievements, a large amount of geological information and data have been accumulated for the description of shale reservoirs, and it has become possible to use machine learning methods for “sweet spots” prediction in shale oil and gas areas. Taking the Duvernay shale oil and gas field in Canada as an example, this paper attempts to build recoverable shale oil and gas reserve prediction models using machine learning methods and geological and development big data, to predict the distribution of recoverable shale oil and gas reserves and provide a basis for well location deployment and engineering modifications. The research results of the machine learning model in this study are as follows: ① Three machine learning methods were applied to build a prediction model and random forest showed the best performance. The R2 values of the built recoverable shale oil and gas reserves prediction models are 0.7894 and 0.8210, respectively, with an accuracy that meets the requirements of production applications; ② The geological main controlling factors for recoverable shale oil and gas reserves in this area are organic matter maturity and total organic carbon (TOC), followed by porosity and effective thickness; the main controlling factor for engineering modifications is the total proppant volume, followed by total stages and horizontal lateral length; ③ The abundance of recoverable shale oil and gas reserves in the central part of the study area is predicted to be relatively high, which makes it a favorable area for future well location deployment.

Shale oil content prediction is currently the focus of exploration. Traditional prediction methods are less accurate, and a large number of analytical tests can increase the cost of exploration. Therefore, new methods need to be introduced to accurately predict sweet spot intervals. In this study, a quick and novel convolutional neural network (CNN) method to predict oil content parameters (TOC, S 1 and S 2) is provided based on well logs (SP, GR, CAL, RT, and AC) in the Dongpu Depression, Bohai Bay basin, China. The results showed that the CNN model showed high accuracy in predicting TOC (R2 = 0.73) than the ΔlogR method (R2 = 0.132). The validation sets for S 1 (R2 = 0.72) and S 2 (R2 = 0.67) also show high prediction accuracy. In addition, Well Wen 318 (located in the same formation) was selected as a blind well to test the performance of the model. The micromigration evaluation method is based on the principle of mass balance, which reflects the changes of the original hydrocarbon generation potential (HGP O) and the real hydrocarbon generation potential (HGP R). Combined with the established CNN model and the micromigration evaluation method, we predicted and evaluated the oil content in lower Mbr 3 of the Shahejie Fm and divided four sweet spots. The differential enrichment of shale oil is controlled by micromigration. Samples with high saturated/aromatic ratio and small main peak carbon of saturated hydrocarbon chromatography are more likely to undergo micromigration, which in turn affects the oil content of shale. Shales with high organic matter abundance and good pore connectivity are more susceptible to micromigration, leading to the differential enrichment of shale oil. This study provides a new idea for evaluating hydrocarbon micromigration of shale oil and predicting favorable zones and the grading evaluation of oil content in petroliferous basins around the world. • A novel CNN method to predict oil content parameters is provided based on well logs. • A micromigration evaluation model was established based on hydrocarbon generation & retention. • 4 sweet spots intervals were predicted combined with micromigration evaluation and CNN model. • The differential enrichment of shale oil is explained from the micromigration perspective. [ABSTRACT FROM AUTHOR]

Continental tight oil sandstone reservoirs are developed in the Cretaceous Quentou Formation in the Songliao Basin, China. At present, there is still a lack of research on the reservoir microstructures, reservoir physical properties, and the division scheme of reservoir types in the Quan 3 Member of the Quanzhou Formation. Therefore, in this paper, taking the Quan 3 Member in the Fuxin Uplift Belt of the Songliao Basin as an example, the microscopic pore structure characteristics of tight oil sandstones have been systematically studied, and the classification standard of tight sandstones has been formulated. Furthermore, the sweet spots of the main production layers are predicted. The results show that the I sandstone group in the Quan 3 Member in the study area belongs to shallow water delta facies. Feldspar lithic fine sandstones are developed in the target layer, and calcareous sandstone is locally developed. Moreover, the mian pore types of the target layer include dissolved intergranular and intragranular pores, followed by primary intergranular pores, while micro-fractures are occasionally seen. According to the mercury intrusion test results, the pore-throat structures of the reservoir in the Quan 3 Member are divided into four types: 1) small-pore medium-throat type (point bar and delta distributary channel), 2) small-pore micro-throat type (point bar), 3) small-pore micro-throat type (natural levee), 4) micro-pore micro-throat type (river floodplain and inter-tributary bay). The lower limits of the physical properties of the effective reservoirs in the Quan 3 Member has been determined: the porosity is 10% and the permeability is 0.1 mD. Finally, combined with the study of the mercury intrusion curves, the physical properties and the sedimentary facies, the classification standard of the Quan 3 Member reservoirs was formulated. For the I sandstone group, the sweet spots of the Type I reservoirs are mainly developed in the Fuyu Oilfield in the southeast areas. The sandstones in this area suffered less compaction, and primary and secondary pores suffered from late dissolution are mainly developed.

Effective shale gas exploration is hindered by the need for obtaining high-resolution correlations between shale strata and the need for classifying shale facies. To address these issues, chemostratigraphy, sequence stratigraphy, and shale gas geology methods were integrated to develop a new method known as "chemical sequence stratigraphy," which was successfully applied to the Wufeng–Lower Longmaxi Formations in the upper Yangtze region. Well Huadi 1 was used as a case study, and detailed data were acquired. Multivariate statistical analyses were applied to three defined indices having different genetic significance, namely: terrigenous input intensity (TII), authigenic precipitation intensity (API), and organic matter adsorption and reduction intensity (OARI). By analyzing the trends of these three indices, the Wufeng–Lower Longmaxi Formations were divided into five fourth-order chemical sequences (from bottom to top): LCW, MCL1-1, MCL1-2, MCL1-3, and MCL1-4. The geochemical facies were named and classified using the chemical sequence stratigraphic framework. The enrichment factor (EF) transformation of elements was conducted to determine whether an element is rich or deficient. The results showed that the favorable geochemical facies in the well were EF-Al deficient, EF-Ca rich, and EF-V rich. The organic matter content and rock brittle strength were then used as chemical parameters, and it was predicted that the LCW and MCL1-1 chemical sequences most likely comprised shale gas sweet spots. This conclusion is consistent with the drilling results and indicates that our proposed method is effective and reliable. This method is further applied to the Changning Shuanghe section, the Shizhu Liutang section, and sections in the Xindi 1 well in the upper Yangtze region. The comparative study of these four sections showed that LCW and MCL1-1 are the key chemical sequences for shale gas exploration and development in the Wufeng–Lower Longmaxi Formations within the Upper Yangtze region. [ABSTRACT FROM AUTHOR]

The oil and gas resources are abundant in the shale of the first member of Qingshankou Formation in Sanzhao Sag, Songliao Basin. To identify the distribution of shale oil sweet spots in the area and guide the deployment of parameter well, the paper studies the sweet spot prediction method based on the logging evaluation of key parameters of single well sweet spot, 3D seismic data and well-seismic combination. First, the cross plot analysis of logging data and core experimental analysis data is used to establish logging evaluation model of lithology, physical properties (porosity and permeability), oil-bearing property (saturation), organic geochemical parameters (TOC and Si), formation pore pressure and compressibility; Then the broadband constrained inversion technology is applied to predict lithology, the prestack elastic parameter inversion to predict physical properties, oil-bearing property and compressibility, and combination of constrained Dix formula inversion and model constrained wave impedance inversion technology to predict formation pressure; Finally, a set of seismic sweet spot identification technology process suitable for the shale oil of Qingshankou Formation in Songliao Basin is established to realize the double sweet spots prediction of geological and engineering sweet spots. The practice shows that the 3D seismic data can effectively predict shale sweet spots, and the actual drilling results are basically consistent with pre-drilling prediction, which proves the effectiveness of technical process. [ABSTRACT FROM AUTHOR]

Effective shale gas exploration is hindered by the need for obtaining high-resolution correlations between shale strata and the need for classifying shale facies. To address these issues, chemostratigraphy, sequence stratigraphy, and shale gas geology methods were integrated to develop a new method known as “chemical sequence stratigraphy,” which was successfully applied to the Wufeng–Lower Longmaxi Formations in the upper Yangtze region. Well Huadi 1 was used as a case study, and detailed data were acquired. Multivariate statistical analyses were applied to three defined indices having different genetic significance, namely: terrigenous input intensity (TII), authigenic precipitation intensity (API), and organic matter adsorption and reduction intensity (OARI). By analyzing the trends of these three indices, the Wufeng–Lower Longmaxi Formations were divided into five fourth-order chemical sequences (from bottom to top): LCW, MCL1-1, MCL1-2, MCL1-3, and MCL1-4. The geochemical facies were named and classified using the chemical sequence stratigraphic framework. The enrichment factor (EF) transformation of elements was conducted to determine whether an element is rich or deficient. The results showed that the favorable geochemical facies in the well were EF-Al deficient, EF-Ca rich, and EF-V rich. The organic matter content and rock brittle strength were then used as chemical parameters, and it was predicted that the LCW and MCL1-1 chemical sequences most likely comprised shale gas sweet spots. This conclusion is consistent with the drilling results and indicates that our proposed method is effective and reliable. This method is further applied to the Changning Shuanghe section, the Shizhu Liutang section, and sections in the Xindi 1 well in the upper Yangtze region. The comparative study of these four sections showed that LCW and MCL1-1 are the key chemical sequences for shale gas exploration and development in the Wufeng–Lower Longmaxi Formations within the Upper Yangtze region.

Abstract Shale reservoirs are characterized by low porosity and strong anisotropy. Conventional geophysical methods are far from perfect when it comes to the prediction of shale sweet spot locations, and even less reliable when attempting to delineate unconventional features of shale oil and gas. Based on some mathematical algorithms such as fuzzy mathematics, machine learning and multiple regression analysis, an effective workflow is proposed to allow intelligent prediction of sweet spots and comprehensive quantitative characterization of shale oil and gas reservoirs. This workflow can effectively combine multi-scale and multi-disciplinary data such as geology, well drilling, logging and seismic data. Following the maximum subordination and attribute optimization principle, we establish a machine learning model by adopting the support vector machine method to arrive at multi-attribute prediction of reservoir sweet spot location. Additionally, multiple regression analysis technology is applied to quantitatively predict a number of sweet spot attributes. The practical application of these methods to areas of interest shows high accuracy of sweet spot prediction, indicating that it is a good approach for describing the distribution of high-quality regions within shale reservoirs. Based on these sweet spot attributes, quantitative characterization of unconventional reservoirs can provide a reliable evaluation of shale reservoir potential.

The Upper Ordovician Wufeng Formation (O3w)-Lower Silurian Longmaxi Formation (S1l) in the southeastern part of the Sichuan Basin is a key field of current shale gas exploration in China. However, there are few studies on "geological sweet spots" prediction based on quantitative characterization of shale gas sedimentary facies. In this paper, taking the Dingshan area in the southeastern Sichuan Basin as an example, the sedimentary facies of shale and its control on the distribution of geological sweet spots in this area were systematically analyzed using the method of point (sedimentary facies analysis using cores) to line (sedimentary facies analysis using single well) to plane (planar sedimentary facies analysis), combined with shale sedimentary background analysis, logging and seismic data. The results showed that, the Wufeng Formation-Longmaxi Formation in the Dingshan area mainly developed with carbonaceous siliceous shale, carbonaceous shale, calcareous shale, silty shale, argillaceous siltstone and calcareous siltstone. The Wufeng Formation and the I subsegment of the Long 1 member are the high-quality shale gas intervals. The high-quality shale section of the target layer in the Dingshan area has high TOC content, good gas content and shale thickness of more than 30 meters, which is conducive for shale gas development. Through the analysis of sedimentary environment recovery and sedimentary facies, it was confirmed that there was an underwater uplift in the Late Ordovician-Early Silurian near the Dingshan 1 well, which controlled the distribution of sedimentary facies and the development of high-quality shale, and finally plays a decisive role on the distribution of geological sweet spots. [ABSTRACT FROM AUTHOR]

Shale reservoirs are characterized by low porosity and strong anisotropy. Conventional geophysical methods are far from perfect when it comes to the prediction of shale sweet spot locations, and even less reliable when attempting to delineate unconventional features of shale oil and gas. Based on some mathematical algorithms such as fuzzy mathematics, machine learning and multiple regression analysis, an effective workflow is proposed to allow intelligent prediction of sweet spots and comprehensive quantitative characterization of shale oil and gas reservoirs. This workflow can effectively combine multi-scale and multi-disciplinary data such as geology, well drilling, logging and seismic data. Following the maximum subordination and attribute optimization principle, we establish a machine learning model by adopting the support vector machine method to arrive at multi-attribute prediction of reservoir sweet spot location. Additionally, multiple regression analysis technology is applied to quantitatively predict a number of sweet spot attributes. The practical application of these methods to areas of interest shows high accuracy of sweet spot prediction, indicating that it is a good approach for describing the distribution of high-quality regions within shale reservoirs. Based on these sweet spot attributes, quantitative characterization of unconventional reservoirs can provide a reliable evaluation of shale reservoir potential. [ABSTRACT FROM AUTHOR]

The accumulation pattern of the marine shale gas in South China is different from that in North America. The former has generally thin reservoirs and complex preservation conditions, so it is difficult to make a fine description of the structural features of shale formations and to reflect accurately the distribution pattern of high-quality shale by using the conventional 2D and 3D seismic exploration technology, which has an adverse effect on the successful deployment of horizontal wells. In view of this, high-precision 3D seismic prospecting focusing on lithological survey was implemented to make an accurate description of the distribution of shale gas sweet spots so that commercial shale gas production can be obtained. Therefore, due to the complex seismic geological condition of Jiaoshiba area in Fuling, SE Sichuan Basin, the observation system of high-precision 3D seismic acquisition should have such features as wide-azimuth angles, small trace intervals, high folds, uniform vertical and horizontal coverage and long spread to meet the needs of the shale gas exploration in terms of structural interpretation, lithological interpretation and fracture prediction. Based on this idea, the first implemented high-precision 3D seismic exploration project in Jiaoshiba area played an important role in the discovery of the large Jiaoshiba shale gas field. Considering that the high-quality marine shale in the Sichuan Basin shows the characteristics of multi-layer development from the Silurian system to the Cambrian system, the strategy of shale gas stereoscopic exploration should be implemented to fully obtain the oil and gas information of the shallow, medium and deep strata from the high-precision 3D seismic data, and ultimately to expand the prospecting achievements in an all-round way to balance the high upstream exploration cost, and to continue to push the efficient shale gas exploration and development process in China.

With the continuous improvement of exploration and the continuous application of new technology, deep coal seam has become an important target. The seismic facies controlled inversion can provide a low frequency background constraint under the control of seismic facies interface, ensure the reliability of inversion results, and make the inversion result have obvious geological significance. The deep coal reservoir of Yushe Wuxiang block of Qinshui Basin has thin coal seams low signal-to-noise ratio, and is difficult to be described. Using the seismic facies controlled inversion, where the detailed division of sedimentary facies, reservoir physical property analysis, reservoir forming conditions and well log data were made as constrains, the “sweet spot” distribution of the Yushe Wuxiang block was predicted. The sweet spot was distributed at the middle and north of study area. The prediction results show that the inversion results can be used to predict the development characteristics of coal seam, and Lame coefficient multiplied by density can be an indicator to predict the sweet spot, which significantly improves the accuracy of sweet spot prediction for the deep coal of the Yushe Wuxiang block of Qinshui Basin. [ABSTRACT FROM AUTHOR]